We are witnessing the rise of a novel class of wearable devices equipped with various sensing capabilities as well as further miniaturization of sensing components that are nowadays being integrated into mobile devices. The inherent mobility of such devices has the capacity to produce dense and rich spatiotemporal information about our environment creating the mobile Internet of Things (IoT). The management of mobile resources to enable sensor discovery and seamless integration of mobile geotagged sensor data with cloud-based IoT platforms creates new challenges due to device dynamicity, energy constraints, and varying sensor data quality. The paper presents an ecosystem for mobile crowdsensing applications which relies on the CloUd-based PUblish/Subscribe middleware (CUPUS) to acquire sensor data from mobile devices in a context-aware and energy-efficient manner. The ecosystem offers the means for location management of mobile Internet-connected objects and adaptive data acquisition from such devices. In addition, our solution enables filtering of sensor data on mobile devices in the proximity of a data producer prior to its transmission into the cloud. Thus it reduces both the network traffic and energy consumption on mobile devices. We evaluate the performance of our mobile CUPUS application to investigate its performance on mobile phones in terms of scalability and CPU, memory and energy consumption under high publishing load.